© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 86357
    Prev
    Next
    INDEX
    Explanations

    actions and their direct objects

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on present‐tense verbs (and their immediately surrounding function words) that describe purposes, capabilities, or requirements (e.g. “provides,” “allows,” “requires,” “demonstrates,” “guide,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     as
    -1.51
     until
    -1.30
     such
    -1.29
     the
    -1.17
    igheid
    -1.14
      
    -1.09
    нем
    -1.08
     écrivain
    -1.08
     there
    -1.04
     هناك
    -1.04
    POSITIVE LOGITS
    
    1.13
     fatta
    1.12
     белые
    1.12
     gefeiert
    1.12
     большие
    1.10
     permitiendo
    1.08
    ు
    1.08
    fting
    1.08
     хорошие
    1.07
     commandé
    1.07
    Activations Density 0.132%

    No Known Activations